Repart Noд6229-LA Govemment Spending оп Health in Lao PDR: Evidence апд Issues Deremбer 20ц ® Бtапдагд Diьclalmeг ]'his vлШте is а ргодие rtf Ие sиК пf Ме Iпмлаппюl Вапk bг Rегоиьo-игъпп апд Developmmt/Ие Wпдд вмk Thе Ипдиgе, пкryreтпопг, мд тrclшШт енргvsнllп Ф1s рарп дп м[ле®aNy N1ct tFevlews N[Ие Етгмvе Dlгхюrs N[Ие Wodd 9м4 пг Ие gоvетпепы Феу rcргеят[. iИе WоЛд 9вп4 дпп ro� guaraпree the кглгагv о1 гИе tlata iпгlидед iп this ногk ТИе Ьолпдвпеч, mloq деюпlпвИтis,тд o�her'тfппиНопчhпип ппanу парИ[Иивvтk до ю[мрlуапу)одgтел[ м �hepart rtf the Wm1d Вап4тпгегпiпg Ие 1ega1 г�атг п1 апv мi�пгу пг �М тдпыетем пг апер�алге о1 sлгИ Ьыпдагiы. Фpytlgh[61a1mieW: SM пв�егfаtlп[Юь рлЫlпИм цтрулgМ Э. Слрушg алд/ог o-ппsпдмпg рпгьоюогаll rtf [Мь wmk W[Ипи[ ректlsтп тву Ье virtlmiм о( applicaбe 1aw. The 1п[епаИопаl 9вп4 fпг Rпопчмааiпп апд �evelopmem/Ие Wпг1д Папk ептигаgеа дUsятlпаепппfця workand иФ1 гшппаllу greы реппiьчlоп ю reргодиreрпгtlпптf tlю wmkprvrtptly_ For реппlsсИп �п рМмору пг герпи му рвп of tlш wrtгq рlеаяе гепд в геqлv[ wtN mrtplere LдппюИм т �Ие Фр[пgИ[ С1еыт:ге СтМ, 1пс, 2II RrtsewrtW Огbе. �атеп, МА П1ИЭ, ISA, R1ерИме W&Тй13ЭФ, 1ы W&]6➢h99o, h�N:/ /иww тоегiеИстт/. А11 пйа qлепеs пп пghтапд @тгы., Игlлдlпg паЬыдLиу пghm, sИолlд Ье аддгеяед �п Фе � е of еЬе РиЫlйн, Thе Wodd 9мk,1818 Н Sreд NW, Wыh'mgЮryOC LЮ33,1]5A,6sZD2��2�2422,e'rtail poбrlghlь'е]ьvтlдЬалkоау_ GOVERNMENT SPENDING ON HEALTH IN LAO PDR: EVIDENCE AND ISSUES' Executive Summary This policy note on government spending for health in the Lao People's Democratic Republic (Lao PDR) provides an overview of government budgetary health spending, primarily covering fiscal years (FY) 2005/06 to 2011/12. The note analyzes overall trends in government health financing and expenditure patterns and discusses some of the efficiency and equity issues pertaining to current government health spending patterns. The policy note is one of a series of health financing analyses, complementing earlier policy notes focusing on out-of-pocket spending as well as community-based and social health insurance schemes in the country. Despite notable progress in health on some fronts, Lao PDR continues to have some of the worst maternal and child outcome indicators, both globally as well as in the East Asia and Pacific region. From a health financing perspective, out-of-pocket payments are the largest share (46.4%) of total health expenditure in the country. This reliance on out-of-pocket payments represents a considerable financial barrier to utilization of health services. The prominence of out-of-pocket spending in the form of user fees and revolving drug funds (RDFs) also raises concerns over management of funds at health facility level and regarding the potential for over prescription. In contrast, social health insurance expenditures are very low in Lao PDR: social insurance schemes cover about 11.4 percent of the population but account for only about 2.8 percent of total health spending. By international standards, government spending on health - at 1.1 percent of GDP - is low and has been erratic over time, primarily due to external financing; domestically-financed government health spending has averaged about 0.5 percent of GDP in recent years. Health's share of the government budget was about 4.3 percent in Fiscal Year 2009/10, and has been increasing. The Lao government has committed to increasing government spending to 9 percent of the budget, implying roughly a three-fold rise compared to planned spending for Fiscal Year 2011/12. If the policy goal is to raise government health spending equitably across the provinces, achieving it will be challenging. Some provinces, notably the more heavily populated ones, will have large expenditure gaps to bridge. There are large disparities in government health spending between central and sub-national levels. In Fiscal Year 2009/10, less than a third of government health spending occurred at the provincial level. Furthermore, provinces differ in the share of total government spending dedicated to health: provinces with higher levels of overall government spending per capital also tended to spend more on health. Per capita health spending was higher in more sparsely populated and poorer provinces such as Sekong and Attapeu as opposed to Champassak, Savarmakhet, and Vientiane Capital. Budgetary provincial spending rates, however, may not give an accurate picture of equity because of cross-provincial utilization, especially at higher levels of care. In addition, hospitals in more affluent provinces/ cities are likely to raise a larger share of their review from user fees and drug sales. 2 Although the recently-announced planned increases in government health spending are welcome, challenges remain. These include ensuring that the additional resources are used to improve access to and utilization and quality of health services - especially in more remote areas - and progressively making additional domestically-financed resources available to reduce both dependence on external funding and out-of-pocket spending for health. To attain these objectives, the government should consider an appropriate mix of both demand-side and supply-side incentives. Instead of, or in addition to, setting a target for budgetary outlays for health, the government needs to improve the efficiency of existing outlays, the measurement of which requires the monitoring of key population health outputs. These should include focus on the level and equity of basic immunization rates, of skilled birth attendance, of institutional delivery rates, of need-based outpatient and inpatient utilization rates, and on adequate levels of financial protection from adverse health shocks. The planned implementation of the free maternal and child health policy is a welcome step in the right direction. However, implementation will need to be complemented by improvements in the capacity of health facilities, not just in clinical and service availability terms, but also in terms of their ability to manage and allocate revenues appropriately. Current weaknesses include inconsistent implementation of user fee regulations and revenue management, variation in management practices, weak procurement practices for drugs, and inadequate service provision levels. In addition, the planned removal of user fees, as envisioned under the free maternal and child health policy, may not be sufficient to improve utilization and inequalities across the country. To achieve this, the government should consider additional demand-side incentives, especially in rural areas. Budget norms, if used prudently, can be a useful framework for determining allocations for all expenditure types. For capital investments, population distribution and needs should be considered. For recurrent costs, equity adjustments could help reduce vertical inequality in personnel budgets. The same principles could be applied to allocation of non-wage recurrent expenditure across provinces and districts, taking into consideration not just population size but capacity and health-sector needs. The government should also consider the capacity of health facilities to generate revenue through state health financing and insurance schemes, and provide clear guidance on which types of expenditure are to be funded by non-wage recurrent budgetary allocations and which are to be funded by user fees. Various unresolved questions remain, including on performance incentives at hospitals, on reforms to provider payment mechanisms, and on drug procurement practices. In conjunction with other reforms, use of informed budget norms will be necessary to achieve efficient and equitable use of resources as the government proceeds with plans to increase its spending on health to 9 percent. 1 This policy note is based on a report prepared by Magnus Lindelow and Laurence Lannes, with input from Loraine Hawkins, Phetdara Chanthala, Sophavanh Thitsy and Robert McLaughlin. In order to address recent developments and to address peer reviewer comments, additional data analysis and text have been added by Wei Aun Yap, Rong Li and Ajay Tandon. Peer review comments were received from Paolo Belli, Antonio Giufdrida, and Christoph Kurowski. The team is also grateful for input and suggestions from Chansaly Phommavong, Vanxay Souvannamethy, Phadeumphone Sonthani and Souphab Panyakeo. 3 I. Introduction 1. In the years following establishment of the Lao PDR, health care was primarily funded through the government budget, with financial and in-kind support from the Soviet Union, Vietnam, and China. A limited range of services was provided for free through a small -- albeit expanding -- network of government facilities. Over time, constrained budget resources and declining support from traditional partners led to a growing reliance on households and communities to finance health care. Initially, this took the form of RDFs, which were established on a pilot basis in the early 1990s. In 1995 a national policy on user fees was officially adopted. This in turn has led to the gradual establishment and expansion of four different health insurance and safety net programs: (i) the Civil Service Health Insurance, (ii) the Social Health Insurance Fund, (iii) Community Based Health Insurance, and (iv) Health Equity Funds. Over the same period, the health sector has seen a rapid expansion of development assistance from new partners, who are financing a wide range of investments and activities. Nonetheless, the coverage of health insurance schemes remains low, and out-of-pocket (OOP) payments are currently the largest source of financing for health in Lao PDR. 2. The high reliance on OOP spending to finance health care has significant implications for both access and equity. This is a growing concern for the government and the Seventh Health Sector Development Plan (2011-2015) highlights "sustainable health financing" as a priority area. The plan calls for an increased government health budget, expansion of pre- payment schemes, and development of mechanisms to ensure that the poor have access to health services. Given the inherent challenges in expanding pre-payment schemes in Lao PDR, which has high levels of poverty and a largely informal labor market, government spending will need to play a key role in expanding access, either by directly financing the costs of health care, or by subsidizing health insurance in some way. One concrete step in this direction is the recently-announced policy to make maternal and child health services free to a large segment of the population. 3. The overall economic outlook for Lao PDR is positive. Economic growth is projected to be 8.3 percent in 2012, and is expected to be in the range of 7-8 percent over the period 2013- 2015.2 Government revenues have risen significantly in recent years due to growth in natural resource revenues, improved tax administration and other factors. However, government spending has also risen at the same time, and at a faster pace.3 As a result, this year's fiscal deficit is projected to increase to 3.2 percent of GDP, up from 2.7 percent the previous year.4 Although Lao PDR's risk of debt distress remains high, most debt indicators show improvement and the country has a relatively comfortable level of debt-service ratio due to the concessions often attached to official borrowing.5 4. Before considering any possible expansion of government spending on health, however, it is important to start from a clear picture of current financing and expenditure patterns, with full appreciation of the efficiency and equity issues pertaining to current spending patterns. With this in mind, this note provides an overview of current financing arrangements in the health sector, focusing specifically on government budgetary health spending. The note reviews and summarizes available data from the Official Gazette and other sources, and presents new evidence on facility revenues and other topics collected as part of the study. This note is one of a series of complementary health financing analyses on out-of-pocket spending and community-based and social health insurance schemes in the country.6,7,8 Additional analytical work in progress will review and assess demand-side pilot interventions currently being initiated by the government, such as the national free maternal and child health policy and the conditional cash transfer pilot. The remainder of the paper is organized as follows. Section II provides a brief overview of Lao PDR's health outcomes and its health system. Section III summarizes the health financing situation. Section IV focuses on government expenditures for health. Section V concludes with a discussion of some key policy issues related to government health spending in the country. 2 World Bank (2012), Lao PDR Economic Monitor May Update, Vientiane: World Bank. 3 This increase has been driven mainly by growth in the wage bill, which in turn is due to an increase in the salary multiplier from 1,500 to 3,000 over the period, expanded staffing quotas, and the upgrading of staff educational accreditation. For further details, see World Bank (2011), Budget Brieffor Lao PDR, Vientiane: World Bank. 4 World Bank (2012). 5 World Bank (2012). 6World Bank (2010), Out-of-pocket Spending and Health Service Utilization in Lao PDR: Evidence from the Lao Expenditure and Consumption Surveys, Vientiane: World Bank. Available from: http://www- wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2010/12/22/000356161_20101222 022357/Rendered/PDF/58579WPOLECS010BOX353808BO1PUBLIC1.pdf 7 World Bank (2010), Community-Based Health Insurance in Lao PDR: Understanding Enrollment and Impacts, Vientiane: World Bank. Available from: http://www- wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2010/12/22/000334955_20101222 023815/Rendered/PDF/585770WPOCBHI01BOX353808BO1Englishl.pdf 8 World Bank (2010), Enrollment of Firms in Social Security in Lao PDR: Perspectives from the Private Sector, Vientiane: World Bank. Available from: http://www- wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2010/12/22/000333038_20101222 021203/Rendered/PDF/585800WPOSSOOplO1EnglishlOlPUBLIC1.pdf 5 II. Health Outcomes and Health System Overview 5. Lao PDR is a relatively small, sparsely-populated and landlocked country surrounded by Cambodia, China, Myanmar, Thailand and Vietnam. Lao PDR had a total estimated population of 6.3 million in 2011. With a GNI per capita of US$1,130 in 2011, it is one of the poorest countries in the East Asia and Pacific (EAP) region. Almost 30 percent of the population live below the national poverty line. Two-thirds of the total population reside in rural areas. Education levels remain low and nearly a third of people aged 15 and above are illiterate. A large proportion of the working population is self-employed (88 percent).9 In spite of these challenges, the government has made continued progress in socio-economic development to accomplish the objectives and goals in the Seventh National Socioeconomic Development Plan. 6. Lao PDR has made steady and significant progress on several key population health outcomes over the past few decades. Life expectancy has increased steadily to over 67 years in 2011, up from about 46 years in 1970. Under-five and infant mortality rates have also declined significantly over the same period. The under-five mortality rate, for instance, has moved from 214 per 1,000 live births in 1970 to 42 per 1,000 live births in 2011 (Figure 1). Under-five and infant mortality rates are above average relative to GNI per capita in this newly reclassified lower-middle income country. At current trends, Lao PDR is projected to meet the child-health Millennium Development Goal (MDG) which calls for a two-thirds reduction in under-five mortality over the period 1990-2015. Lao PDR appears also to be on- track to attain the maternal health MDG (which calls for a 75 percent reduction in the maternal mortality ratio over 1990-2015), although there remain some concerns regarding the accuracy of maternal mortality estimates. Figure 1: Trends in key population health outcomes in Lao PDR, 1970-2011 Population health indicators for Lao PDR, 1970-2011 U)0 0 Under-five mortality 0 a- Infant mortality 0 Lie xpcany2 0 19a rYea SPlo L io ab a) a) _J C' Life expectancy - 1970 1980 1990 2000 2010 Year Source: WDI Note: y-scale logged 7. Despite notable progress in health on some fronts, considerable challenges remain. While on-track for the child-health MDG, and by some estimates for the maternal-health MDG, Lao PDR continues to have some of the worst maternal and child health (MCH) outcome indicators, both globally as well as in the EAP region. As mentioned above, although there is some uncertainty regarding exact numerical values, WHO/UNICEF/UNFPA/World Bank (2012) estimate that, at 470 per 100,000 live births, Lao PDR's maternal mortality ratio (MMR) is among the highest in the world, equal almost to double that of neighboring Cambodia and to eight times that of Vietnam (Table 1).1o About a third of all children under five remain underweight. At current trends, Lao PDR is off-track on the nutrition MDGs, and there are significant urban-rural, geographic, and ethnic-group related inequalities in health outputs and outcomes. Underlying poor MCH outcomes include low levels of coverage for key MCH utilization indicators such as antenatal care (ANC), skilled birth attendance, and immunization for measles and DPT. Communicable conditions continue to dominate the overall disease burden in the country (Table 1). Table 1: Key population health indicators for Lao PDR and comparators GNI Maternal Skilled birth Percent Neonatal Under- DPT Measles Malnutrition Proportion of per mortality attendance with at mortality five coverage coverage prevalence mortality due to capita, ratio rate (%) least 1 rate mortality (%) (%) among under- communicable US$ ANC visit rate fives (weight- diseases (%) for-age) Bhutan $2070 180 64.5% 74.4% 26 55.9 91% 95% 12.7% 53% Cambodia $830 250 71% 89.1% 22 46 92% 93% 29% 60% China $4940 37 99.3% 94.1% 11 15.9 99% 99% 3.4% 15% India $1410 200 52.7% 75.1% 32 63.4 72% 74% 43.5% 52% Indonesia $2940 220 82.2% 93.3% 17 33.3 83% 89% 19.6% 41% Lao PDR $1130 470 37% 37% 21 43.9 74% 64% 31.6% 58% Nepal $540 170 18.7% 58.3% 28 50.3 82% 86% 29.1% 60% Sri Lanka $2580 35 98.6% 99.4% 10 12.6 99% 99% 21.6% 11% Thailand $4420 48 99.4% 99.1% 8 12.8 99% 98% 7% 24% Timor- $2730 300 29.3% 84.4% 24 57.6 72% 66% 45.3% 76% Leste Low $552 452 52% 79.1% 32.6 104 79.5% 77.2% 21.7% 71.4% income Lower $2357 213 75% 84.2% 20.3 51.1 87.2% 85.7% 15.3% 45.5% middle income East Asia $3252 138 80% 85.5% 14.4 30.9 85.6% 83.9% 18.3% 37.7% & Pacific Source: WHO & WDI; Note: Data are for latest available year, 2006-2011 8. There are large inequalities in the utilization of health services by socioeconomic status. Individuals in the richest quintile of the population spent five times more on out-of-pocket healthcare payment as a share of total household consumption than those in the poorest quintile in 2007-2008 (Figure 2). Table 2 also shows that the utilization of both inpatient and outpatient services are greater among the rich individuals. The widening in this inequality might reflect a shift towards higher quality / more expensive health services by the rich and an increase in financial protection for the poorest groups. In addition, there is a startling variation in health indicators across provinces in Lao PDR. In some provinces such as Vientiane Capital and Luang Prabang, the level of DPT3 immunization is even comparable to those of developed countries, suggesting that there may be lessons to be learnt from better-performing provinces (Figure 3). Figure 2: Out-of-pocket healthcare payments as share of total household consumption, 2007-2008 ? The financial burden of health care payments 00 0 0 Poorest Q2 Q3 Q4 Richest Source: Lao Expenditure and Consumption Survey in 2007-2008 Table 2: Proportion of individuals who use health service by quintile, 2007-2008 Outpatient care sought (%) in the past 4 weeks Inpatient care sought (%) in the past year Poorest 1.0% 1.1% Q2 1.3% 1.5% Q3 1.7% 1.6% Q4 1.9% 2.1% Richest 2.8% 2.6% Source: Lao Expenditure and Consumption Survey in 2007-2008 Figure 3: Health indicators for Lao provinces compared with selected countries, 2010 Global comparison of Lao PDR provinces, 2010 DPT3 immunization Skilled birth attendance China 0 China Loangprabang Germany Vientiane capital vI r!..opl Denmark Vietnam Vientiane capita emakwia Oudommndonesia Oudomxay Indonesia o> Saravanh Lao PDR 0 Savannakhet Huaphan Liberia pia New Gurne E eUganda Bokeo Papua New Guine Savannakhet Liria Boakhamly i 0 Borikhamxay La. PDR Somalia Somai a Phongaiy Loangprabang d EquatoIal Guinea Huaphan Niger Oudomxay - __________ __________I ____-_____ Ethiopia Lao PDR Other countries Lao PDR Other countries Source: WHO and National Health Statistic Report, 2010 9. Health care in Lao PDR is largely delivered by the public sector. There are a few private clinics, primarily in urban areas. At 13 doctors, nurses, and midwives per 10,000, human resources for health (HRH) workers are far below the WHO recommendation of 23 HRH workers per 10,000 people needed for minimal coverage of key health system interventions. At 7 per 10,000, in-patient bed density in the country is also lower than that of other low and lower-middle income countries. Other challenges facing the Lao health system include poor service readiness, low quality of health care, and weak planning, financial management, implementation and monitoring/reporting capacity, especially at local levels, but also at higher levels. 9 World Bank (2012), World Development Indicators, Washington DC: World Bank. 1o WHO/UNICEF/UNFPA/World Bank (2012), Trends in Maternal Mortality: 1990-2010, Geneva: World Health Organization. III. Health Financing Summary for Lao PDR 10. Total health expenditure in Fiscal Year 2009/10 was estimated to be KN 1,399 billion (Lao kip), about 2.5 percent of GDP and roughly $27 per capita.11 Financing for the health sector in the country comes from three primary sources: (i) OOP payments by households; (ii) the government budget; and (iii) external resources from donors and projects (Figure 4). Recent National Health Accounts (NHA) estimates indicate that, at 46.4 percent, OOP payments account for the largest share of total health expenditure in the country. Government health spending, which includes external financing, comprised about 40.7 percent. The external financing accounts for 31.9 percent of total health spending. The contribution of social health insurance schemes was only about 2.8 percent of total health spending in Fiscal Year 2009/10.12 Figure 4: Health financing flows in the Lao health system Private providers Direct payments for fees and medicines population Saare Salaries Investment Budqet Recurr. cos6 Hospital Revenues Budget MOH / HpA Province External Donors, projects Donors, projects assistance Civil serv. Contributions L4. Payment Formal Fund / sector scheme 11. The heavy reliance on OOP payments in Lao PDR results in considerable financial barriers to utilization of health services, contributing to low levels of utilization and significant health-related financial risk. Contributory health insurance schemes can play a role in addressing these challenges. However, with the majority of the population engaged in subsistence farming or working in the informal sector, expanding the coverage of such schemes is likely to remain difficult in the foreseeable future. As underscored in the government's own draft Health Financing Strategy for 2011-2015, increasing and improving the effectiveness of existing government health spending is likely to be the most viable strategy for improving access to health care services and enhancing financial protection.13 Given this, the remainder of this paper focuses specifically on analyzing Lao government spending on health. 11 Ministry of Health & WHO (2012), Lao PDR National Health Accounts FY 2009-2010, Vientiane: Ministry of Health, Lao PDR. 12 Although informative, the NHA data needs to be interpreted with caution. NHA estimates typically do not manage to capture all external financing due to fragmentation, difficulties in defining the scope of support to the health sector, and reluctance of some actors to report their actual spending. Evidence also shows that NHA estimates often underestimate private out-of-pocket spending. This is likely to be the case in Lao PDR, as spending estimates are based on a household survey with a weak health component (Lao Expenditure and Consumption Survey) and data from a study in central hospitals and one provincial hospital. 13 Ministry of Health (2011), Health Financing Strategy 2011-2015, Draft, Vientiane: Ministry of Health, Lao PDR. IV. Government Spending on Health 12. The remainder of this note focuses on government spending on health, but what exactly does this concept encompass? In the case of Lao PDR, government health spending includes several different elements: (i) budgetary spending by the central Ministry of Health; (ii) budgetary spending at local level (through provincial and district health offices); (iii) budgetary spending by other ministries (e.g. Ministry of Defense); (iv) social security spending by the Social Security Organization (private sector employees) and the State Authority for Social Security (civil servants and employees of state-owned enterprises); and (v) most externally-financed expenditure on health, much of which is not captured in the budget. In addition, expenditure financed by revenues from services and drugs at facility level is sometimes considered government spending, although as discussed in more detail below, this practice is problematic. Budgetary Spending on Health 13. In Fiscal Year 2009/10, based on data in the Official Gazette, government budgetary spending on health at central and local levels was KN 604 billion, roughly 1.1 percent of GDP.14,15 This estimate includes both recurrent and investment (domestic- and externally- financed) expenditure, but does not include health spending by other ministries or spending by social security organizations. Government health spending as a share of GDP has been quite erratic over time, mainly due to fluctuations in external financing of health, which is counted within the investment budget. If only domestically-financed health spending is considered, government health expenditure accounted for 0.5 percent of GDP in Fiscal Year 2009/10, an increase of 0.3 percent from Fiscal Year 2005/06. Planned domestically-financed government spending on health was also 0.5 percent of GDP for Fiscal Year 2010/11 and Fiscal Year 2011/12 (Figure 5). According to the Seventh National Socioeconomic Development Plan, some revenue from the Nam Theun 2 hydropower project is to be allocated to eligible programs as part of the government's budget and preparation process, provided additional external funding can be found16. The state budget plan for Fiscal Year 2011/12 indicates that KN 17 billion of KN 100 billion of government revenue from Nam Theun 2 is to be allocated to the Ministry of Health, with KN 8.1 billion of this allocation intended primarily for demand-side financing initiatives such as health funds and free MCH. It is understood that KN 17 billion has also been allocated for the current Fiscal Year 2012-13 in order to achieve health MDGs 1, 4, and 5, and for poverty reduction. Figure 5: Government spending on health as share of GDP, FY 2005/06-FY 2011/12 Government health spending share of GDP, FY 2005/06-2011/12 - Total government health spending 0- a) a- Domestically-financed government health spending 0- 2005/06 2007/08 2009/10 2011/12 Fiscal year Note: Actual FY 2005/06-2009/10; Planned FY 2010/11-2011/12 Source: Official Gazette & State Budget Plan (various years) 14. In absolute terms, domestically-financed government health expenditure increased from KN 108.1 billion in Fiscal Year 2005/06 to KN 248.4 billion in Fiscal Year 2009/10, representing an average increase of 23.4 percent per year in nominal terms (Figure 6). Planned domestically-financed spending in Fiscal Year 2010/11 was KN 375.6 billion and KN 288.2 billion in Fiscal Year 2011/12. Recurrent spending has generally accounted for 80-90 percent of total government health expenditure, with wages accounting for progressively lower shares of recurrent health spending (from over 80 percent in Fiscal Year 2005/06 to 72 percent in Fiscal Year 2009/10). However, this trend may not hold as the government plans to raise wages and allowances during 2013-2015, with a 35 percent increase in wages for Fiscal Year 2012/13, in an attempt to improve the living standards of civil servants. 17 Figure 6: Domestically-financed government spending on health, FY 2005/06-FY 2011/12 Domestically-financed government health spending trends FY 2005/06-2011/12 Nominal Real Domestic capital a Non-wage recurrent a Wages 2005/06 2007/08 2009/10 2011/12 2005/06 2007/08 2009/10 2011/12 Fiscal year Fiscal year Source: Official Gazette & State Budget Plan (various years) Note: Real terms based on base FY 2009/10 15. Health's share of the total government budget has increased in recent years. In Fiscal Year 2009/10, health expenditure accounted for 4.3 percent of total government expenditure, up from 3.3 percent in Fiscal Year 2005/06. Recurrent health expenditure in Fiscal Year 2009/10 was 3.7 percent of total government recurrent expenditure.18 The government has indicated it will increase government health expenditure to 9 percent of overall government expenditure, although it is not clear whether this target refers to total government health expenditure, domestically-financed government health expenditure, or government health recurrent expenditure. Focusing on total government health expenditure is problematic as this includes external financing, which is changeable and a poor marker of government commitment to health. If the focus is on recurrent government health spending or domestically-financed government health spending, the increase to 9 percent is significant and the feasibility of achieving it will need careful consideration (Figure 7). Figure 7: Government health expenditure share of total government expenditure, FY 2005/06-2011/12 Government health spending Government recurrent health spending share of government spending share of recurrent government spending Planned oActual Actual a a. Planned 2005/06 2007/08 2009/10 2011/12 2005/06 2007/08 2009/10 2011/12 Fiscal year Fiscal year Note: Actual FY 2005/06-2009/10; Planned FY 2010/11-2011/12 Source: Official Gazette & State Budget Plan (various years) Central versus Sub-National Government Health Spending 16. In Fiscal Year 2009/10, less than a third (27 percent) of the government's total (externally- and domestically-sourced) budgetary spending for health occurred at the sub-national (i.e. provincial) level. The share of health spending at provincial level varies by expenditure categories: approximately 65 percent of the government's recurrent health expenditure occurred at the provincial level (primarily for wages) but only 7 percent of capital health expenditure was incurred by provinces. This is possibly due to external capital expenditure being categorized as central expenditure, even if it was spent at the local level. If only domestically-sourced financing is considered, sub-national health spending as a share of total government health spending has been inconsistent over time. For example, in Fiscal Year 2007/08 and 2008/09 up to 94 percent was sub-national, compared with 62 percent in Fiscal Year 2009/10. The share of non-wage recurrent expenditure, however, has been steadily declining (Figure 8). Figure 8: Sub-national domestic government health expenditure as a percentage of total domestic government health expenditure, by expenditure category FY 2005/06-2011/12 Sub-national domestic government health expenditure by spending category F)0 E'a 0 0 ~- CL 45Co 0 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 Total recurrent Wage recurrent Non-wage recurrent Sub-nation over total Note: Actual FY 2006/07-2009/10; Planned FY 2010/11-2011/12 External financing excluded Source: Official Gazette & State Budget Plan (various years) 17. A large part of the Ministry of Health (MOH) central budget is allocated to 25 institutions including dedicated centers, universities, and four central hospitals.19 Budgets to those institutions accounted for about 70 percent of the MOH central budget in 2009/10, with the remainder allocated to ministry departments: Cabinet; Department of Hygiene and Prevention; Department of Curative Care; Food and Drug Department; Organization and Personnel Department; Inspection Department; and Planning and Finance Department. 18. Provinces differ not only in the share of total government spending dedicated to health, but also in the actual amount spent per capita. Not surprisingly, provinces with higher levels of overall government spending per capita also tend to spend more on health (recurrent health expenditure) (Figure 9). Some of this variation appropriately reflects an attempt to address inefficiencies. For example, per capita government spending on health is higher in sparsely populated and poorer provinces such as Sekong and Attapeu, where health needs are likely to be greater, as opposed to in wealthier provinces such as Champassak, Savannakhet, and Vientiane Capital. 20 As expected, there are some correlations between financial inputs (provincial recurrent health expenditure per capita) and density of health workers and hospital beds. (Figure 10).21 Figure 9: Planned provincial recurrent health expenditure per capita, FY 2011/12 Planned provincial recurrent health expenditure per capita, FY 2011/12 Sekong* Sekong* .Loangnamtha Attapeu Loangnamtha Attapeu Lon%amh aO Phon~sI Khammounan Phon saay n* Kham-ouan Xiengkhuange o Xiengkhuangg Bokeo Bokeo XC XC Borhr Borikhamxay 2 HuphanBorrkhrn XY 2 0 Huaphan V tian Bl a Vientiane province% anbHol uap Pan Odmxay a, hatnxy *Saravanh a Saravanh *Champasack *Champasack *Savannakhet > Savannakhet* * Vientiane capital 0 Vientiane capital 400000 800000 1200000 10 20 30 40 50 Provincial govt expenditure per capita (LAK) Provincial poverty rate (%) Source: State Budget Plan & Epprecht (2008) Figure 10: Provincial health workforce and infrastructure by recurrent health expenditure per capita, FY 2010/11 Provincial health workforce and infrastructure by recurrent health expenditure per capita, FY 2010/11 *Vientiane cpital * Vientiane mapital .9 *Attapeu Sekong * 8 - *Xayabouly o Khammouan Loangprabang*e *e Borikhamxay keo a _Phong ly Oudomxay Attapeu Xayabouly W ongsaly Loangpabang . *LoangnM9 *Huaphan * Kham-ouan Vientiane pr.vin-et Xiengldruang I - *CharnpasackOXiengkhuang I Co OudonY * Hphan *ChanpSack Borikhanxay *Sa,Annakhet Sa h entiane province Loangnamtha 10000 40000 80000 10000 40000 80000 Provincial govt reccurent health expenditure per capita (LAK) Provincial govt reccurent health expenditure per capita (LAK) Source:State Budget Plan & National Health Statistic Report (FY 2010/11) 19. For various reasons the above analysis of provincial per capita health expenditure may not provide a fully accurate picture of expenditure equity across the provinces. Firstly, the data is based on spending by providers in each province; however, provinces with regional hospitals also treat patients from neighboring provinces (either by referral or self-referral). Similarly, patients in Vientiane municipality and in nearby provinces tend to use central hospitals for primary, secondary, and tertiary care: these hospitals have higher unit costs. Secondly, there is a lack of reliable information on the costs of providing access to services in remote areas and of meeting higher levels of health need in some areas. Thirdly, hospitals in more affluent provinces/ cities are likely to be able to raise a larger share of their revenue from user fees and drug sales, but data on this is not complete. Finally, some patients from large cities or living near bridges/border crossings may use Thai hospitals. Social Security Spending 20. There are four risk protection schemes in Lao PDR: (i) the State Authority of Social Security (SASS), a mandatory health insurance scheme for civil servants; (ii) the State Security Organization (SSO), a mandatory health insurance scheme for employees from the private sector and state-owned enterprises; (iii) the Community Based Health Insurance (CBHI), a voluntary scheme for the informal sector; and (iv) Health Equity Funds (HEF), which provide financial support to poor households in some areas of the country. Currently only about 11.4 percent of the population is covered by one of these schemes: 6.3 percent is covered by SSAS; 1.5 percent by SSO; 2.1 percent by HEFs and 1.7 percent by CBHI. These figures quantify the low population coverage of these schemes, although the depth of coverage (especially for CBHI) is a further issue.22 21. Despite low coverage, the revenues and expenditures of SSO and SASS have been increasing, and both schemes are operating with a surplus. In 2010, 35 percent of SASS revenues came from civil servant contributions, 44 percent from the government's contribution and 21 percent from last year's balance. As expected, SSO revenues rely more heavily on employees' contributions (74 percent) and the remainder comes from interests earned on savings. Overall, health insurance expenditure from SASS and SSO represents a very small share of GDP: about 0.01 percent for SSO and about 0.03 percent for SASS (Figure 11). The latest available information, for 2011, indicates that SSO expenditure is relatively unchanged at 0.016 percent of GDP, although in absolute terms expenditure has risen by 24 percent.23 Figure 11: SASS and SSO expenditure share of GDP, 2006-2010 SASS and SSO expenditure share of GDP, 2006-2010 SASS C? 0 C N sso a) 2005 2006 2007 2008 2009 2010 2011 Year Source: SASS & SSO 22. The proportion of spending by each scheme on health insurance varies. In the SASS, health insurance represents only 8 percent of total expenditure with long-term benefits accounting for 76 percent, short term benefits 15 percent and administrative costs 1 percent. In the SSO scheme, health insurance accounts for almost half of expenditure (47 percent), short-term benefits 21 percent, long term benefits 9 percent and administrative costs 23 percent of total expenditure.24 23. The number of contributors and beneficiaries (i.e. employees, spouses, children and pensioners) has been growing substantially under both schemes. Total beneficiaries increased from 51,766 in 2005 to 107,737 in 2010 in the SSO scheme (contributors went from 26,374 to 53,075). The rate of increase for SSAS has been faster, with the number of beneficiaries increasing from 98,183 to 317,183 between 2005 and 2010. Per beneficiary health expenditure is significantly different between SSO and SASS but has generally been increasing in both schemes. In 2010, however, per beneficiary SASS expenditure dropped due to an increase of over 300 percent in the number of beneficiaries from 2009. At the same time, nominal expenditure stagnated (Figure 12). Figure 12: SASS and SSO expenditure per beneficiary, 2006-2010 SASS and SSO expenditure per beneficiary, 2006-2010 0 CD 0 ~SSO-s SASS <0 O 0 O (0 'IT 0_ O 0 0 0 2006 2007 2008 2009 2010 Year Source: SASS & SSO External Financing for Health 24. External financing represents a major source of funding for the Lao health sector. However, as off-budget external financing is not well captured, it is difficult to get a clear picture of overall financing. The most recent NHA for 2009/10 show external financing to the health sector to be KN 445 billion (over $54 million), about 32 percent of total health spending in the country. Over the period 2006-2010, an average of 53 percent of disbursed external financing for health went towards health systems strengthening (Figure 13).25 The skew towards health systems - as opposed to vertical disease-specific programs - increased over the period 2007-2010 (Figure 14). Figure 13: ODA disbursements by category, 2007-2010 Official development assistance (ODA) disbursements, 2007-2010 HIV/AIDS Infectious disease control Population Health systems strengthening Source: OECD/DAC Figure 14: ODA commitment and disbursement trends, 2007-2010 ODA commitments ODA disbursements 2007 2008 2009 2010 2007 2008 2009 2010 HIVMAIDS I nfecous diseas con.l J HIAIDS Infectous diseas -nrd Population Healths s emns Population Health eysem Source: OECD/DAC 25. Although the database of the Organization for Economic Co-operation and Development's Development Assistance Committee (OECD/DAC) provides up-to-date data on external financing to the health sector, it is incomplete as it relies on data provided voluntarily by donors.26 This data may underestimate aid flows, both because of under-reporting by donors and the absence of recording of non-governmental entities. Nonetheless, the OECD/DAC and NHA data provide a consistent picture. In 2009-2010, the NHA exercise estimated total external assistance to health at $54 million, while for the calendar year of 2010 OECD/DAC estimated total external disbursements to the Lao health sector at $54 million. Such consistency between the estimates provides confidence in subsequent OECD/DAC estimates for multilateral and bilateral assistance to health. 26. The OECD/DAC estimates highlight three primary characteristics of external financing: (i) external financing flows are substantial in Lao PDR, increasing from $35.8 million in 2006 to $54.2 million in 2010; (ii) there are sometimes large discrepancies between commitments and disbursements; and (iii) multilateral financing for health is increasing relative to bilateral financing (Figure 15). Figure 15: External financing for health in Lao PDR, 2006-2010 External financing for health to Lao PDR, 2006-2010 Multilateral Bilateral Disbursements Committments W 2005 2006 2007 2008 2009 2010 2011 2005 2006 2007 2008 2009 2010 2011 Year Year Source: OECD/DAC International Comparative Perspective 27. By international standards, Lao PDR is characterized by low government spending on health and a relatively high reliance on external assistance. Based on WHO/NHA estimates, general government spending on health (MOH budget and social security expenditures on health) as a share of GDP in 2010 was only 1.5 percent in Lao PDR, against 2.1 percent in Cambodia, 1.8 percent in Nepal, 2.9 percent in Thailand and 2.6 percent in Vietnam (Figure 16).27 At the same time, Lao PDR has one of the highest shares of external finance in total health spending, lower than Cambodia and Timor-Leste but higher than Nepal, Bhutan, and other comparators. International evidence suggests that the government health spending share of GDP would need to increase to lower the OOP share of total health spending, although some countries have reduced the OOP share at relatively low levels of government health spending. Figure 16: International comparison of government health spending and external share, 2010 OOP vs government health spending shares, 2010 o - *Solomon Islands * Vietnam Philippines Lao PR - - Cambodia * Mongolia / * jIndonesia hIna .0 7.Fiji £ 7 £0 Malasnia Mondlia : 1, nnd Mtyi E Vietnam* PNG *Malaysia - 7 * Cambodia 0PNG * Fiji CLa * L PRIndonesia * Thailand - Philippines Solomon Islands * 250 1000 25005000000025000 75000 0 2 4 6 8 10 GDP per capita, US$ Government health spending share of GDP (%) Source: WHO and WDI Revenues from Health Facilities 28. Before 1990, health services were provided free of charge in public hospitals and health centers in Lao PDR. As funds from the budget for non-wage recurrent costs were extremely limited, user fees and RDFs were introduced in the 1990s to improve the quality of services and availability of essential medicines.28 Health facility managers were given flexibility and discretion over use of RDF and user fee revenues. In practice, the importance of user fees and RDF relative to government budget financing and external resources varies significantly from one facility to another. However, the government budget tends primarily to cover salaries, compensation, and allowances, while facilities rely on technical revenues and other revenues for non-wage recurrent expenditure and capital expenditure (and in some cases to payments for contract staff). Moreover, recent reforms in the user fee policy expand the scope of managers and confirm the need of facilities to rely on fees for non-wage recurrent expenditure (Box 1). Box 1. Recent reform of user- fee policy in Lao PDR Between 2005 and 2008, user fee policies and other related policies underwent significant changes through the Curative Law (2005), the State Budget Law (2006), and related decrees and Ministry of Finance (MOF) Instructions, as presented in Appendix 1. The key policy changes were: * Expansion of the scope of user fees to a longer list of medical, training and research services (referred to as "technical revenues"); freedom to charge for non-technical services (e.g. hospital restaurants, leasing of land or buildings); * Health centers can charge user fees; * Instructions on national regulation of fee levels agreed between MOH and MOF. However, in practice, the fee schedule is based on negotiation with central hospitals and a single scale of fees has been issued for all hospitals; * A new Hospital Autonomy policy piloted in Mahosot Hospital since 2009, designed as a means to reduce budget subsidies for salary and non-salary operating costs. These are increasingly to be covered by user fees and RDF revenues; * A shift from supply-side financing exemptions for the poor and other groups towards demand-side reimbursement of user fees from various social health insurance schemes, health equity funds, or government welfare funds. * A Budget Law requirement for hospitals to include planned expenditure financed by technical revenues in their annual budgets, and to return unspent user fee funds to the MOF at the end of the fiscal year. The use of revenues is therefore limited to items that can be financed from the budget and cannot be used for staff incentives. The previous requirement to contribute 20% of revenue to the MOF no longer applies. 29. The Ministry of Health and development partners fund initial RDF stocks to enable drug funds to operate with a surplus. Increasingly, the RDFs are also collecting revenues from health-financing schemes: drug costs are now the largest expenditure category for schemes that reimburse on a fee-for-service basis, accounting for 83 percent of the total value of exemptions for outpatient care and for 65 percent of inpatient exemptions in health equity funds in the southern provinces. 30. RDF regulations and guidelines allow facilities to charge a mark-up of 25 percent on the purchase price of drugs and other commodities to cover the costs of transport, inflation, losses and administration.29 Health facilities can use the surplus - after replenishing stocks, meeting operating and transport costs, and providing free drugs to the poor - for other health facility operating costs but not for the benefit of individuals. In practice, these regulations and guidelines are not well enforced. Box 2 presents evidence from Lao PDR and neighboring countries on the issues related to RDF management. Box 2. Evidence of Poor Management of Revolving Drug Funds Prescribing patterns. Health facilities in Lao PDR are unusually reliant on profit margins from sales of drugs and commodities as a source of revenue. A number of studies have found that this creates incentives for irrational and excessive prescribing.30 Similar patterns of high shares of hospital revenue coming from drug sales have emerged in hospitals in Vietnam and China. To address the unintended consequences, China is now reforming this aspect of its health system by financial and managerial separation of pharmacies from hospitals, together with hospital payment reform.31 Drug procurement and prices. Different assessments in Lao PDR have found that variation in procurement processes and in the application of mark-ups leads to disparities in the prices charged to patients for the same drug. These disparities indicate poor procurement practice, increase the risk of corruption and can lead to poor quality control.32 Procurement is centralized in some provinces and decentralized in others. An assessment of price disparities in facilities in Khammouane, Savannakhet, Champassak and Sekong in May 2011 revealed that prices for the same drugs are higher in provinces with decentralized procurement. It also reveals that drug prices are on average higher in provincial hospitals than in district hospitals and health centers (see Appendix 2 for more details about the assessment, prices and availability of essential drugs). Previous assessments in Lao PDR already highlighted that average mark-ups were higher than the legal 25% (over 40% on average in a 2008-09 study of Vientiane province), principally due to rounding-up of per-tablet prices, but also due to the practice of comparing and adjusting prices in line with private retail pharmacies. 31. Appendix 3 presents the results from a case study conducted in the southern provinces and raises a number of issues in the management of revenues at local level. These include the fact that the government budget almost exclusively finances wages while facilities rely on other revenues for non-wage recurrent expenditures; inconsistent implementation of user fee regulations; inconsistent management of user fee and RDF revenues; variation in management practices for the different health financing schemes; weak procurement practices for drugs; and the difficulty of managing donor/project resources. 14 Note that there is currently no reliable national data on spending by level of care or function in Lao PDR. Additional analysis of spending based on information collected during field visits to selected provinces, districts, and facilities is provided in Appendix 3. Ongoing policy discussions with the government through the Poverty Reduction Support Operation (PRSO) envision the adoption of organization codes for Tier-2 budget entities, which will allow further disaggregation in the future. 15 Ministry of Finance (2011), Official Gazette: State Budget Revenue-Expenditure Implementation of FY 2009- 2010, Vientiane: Ministry of Finance. 16 An in-depth review of the allocation of NT2 revenues is beyond the scope of this PER, but can be perused from the cross-sectoral Lao PDR Public Expenditure Review (2011), World Bank. 17 Government of Lao PDR (2012). Decree No 221, 30 May 2012 18 In the budget of the Ministry of Health, all recurrent expenditures are recorded as domestically financed expenditure (unlike for capital expenditures); it is therefore impossible to disentangle domestic and external recurrent spending on health. 19 These institutions include the National Centre for Information and Education for Health; Malaria, Parasitology and Entomology Centre; National Centre for Laboratory and Epidemiology; National Centre for Environment Health and Water Supply; Maternal and Child Health Center; National Tuberculosis Center; Dermatology Center; Rehabilitation Center; Ophthalmology Center; Medical Supply Center; Food and Drug Analysis Center; Traditional Medicine Center; Center for HIV/AIDS & STI; National Institute of Public Health; Mahosot Hospital; Mittaphab Hospital; Mother and Child Hospital; Setthathirath Hospital; University of Health Sciences; Public Health & Medical Science College; Luang Prabang Health College; Savannakhet Health College; and Champassak Health College. 20 Poverty data are from 2008; See Epprecht, M, N Minot, R Dewina, P Messerli, A Heinimann (2008), "The Geography of Poverty and Inequality in the Lao PDR". Swiss National Center of Competence in Research (NCCR) North-South, University of Bern and International Food and Policy Research Institute (IFPRI), Bern: Geographica Bernensia. 21 Similar correlations are observed by provincial total health expenditure per capita. 22 Detailed examination of these health financing schemes is beyond the scope of this PER. Recent analyses by the World Bank address these schemes in greater detail: World Bank (2010), "Community-Based Health Insurance in Lao PDR: Understanding Enrollment and Impacts," Vientiane: World Bank. World Bank (2010), "Enrollment of Firms in Social Security in Lao PDR: Perspectives from the Private Sector," Vientiane: World Bank. 23 Data for SASS for 2011 is unavailable at time of revision. 24 In both schemes, long-term benefits encompass pensions, child and disability allowances; short-term benefits include pregnancy and death allowances (and sick leave in the case of SSO). 25 Based on health spending data from the OECD/DAC Credit Reporting System and categorization of Shiffman, J, D Berlan, and T Hafner (2009), "Has Aid for AIDS Raised All Health Funding Boats?" Journal of Acquired Immune Deficiency Syndromes, 52: S45. 26 http://stats.oecd.org/Index.aspx?DatasetCode=CRSNEW [accessed on Sept 26, 2012] 27 According to data presented in Figure 13, our estimates of government spending as a share of GDP using official data from the government are lower, only 1.1% of GDP. This is likely due to differences in estimation methods and GDP source data. 28 Nationwide implementation of these policies was formalized in 1995 (Decree 52/PM on user fees; RDF regulations No.2635). 29 The RDF decree and guidelines specify processes and practices for good management of drug procurement, distribution, inventory management, rational use of medicines, financial management, and charges to patients. 3oKeohavong, B, L Syhakhang, S Sengaloundeth, A Nishimura, K Ito (2006), "Rational use of drugs: prescribing and dispensing practices at public health facilities in Lao PDR", Pharmacoepidemiology and Drug Safety, 15; Murakami, H, B Phommasack, R Oula, S Sinxomphou (2001), "Revolving drug funds at front-line health facilities in Vientiane, Lao PDR", Health Policy and Planning, 16; Ron, A, B Phonvisay, K Manivong, B Jacobs (2010), "Medicine prescribing and the provider payment method in community- based health insurance in Lao PDR", Essential Medicines Monitor, Geneva: World Health Organization. 31 Langenbrunner, J (2010), Fixing the Public Hospital System. China Health Policy Notes, No. 2. Washington DC: World Bank. 32 Savedoff, W (2008), "The impact of information and accountability on hospital procurement corruption in Argentina and Bolivia", U4 Brief, 7, USA: Chr. Michelsen Institute. V. Discussion Attaining the 9 Percent Target 32. The Lao government is committed to increasing its spending on health to 9 percent of the state budget. This review does not analyze the capacity of the health system to absorb this additional revenue, nor comments on motivations for this commitment, but seeks to answer the question of what it will take to reach this target. Figure 17 shows the gap given current and planned levels of spending at both the central and sub-national levels with separate charts for total and recurrent spending. The gap is particularly large at central level for both total and recurrent health spending. In either case, attaining the 9 percent target will require a spending increase of roughly 300 percent. In the absence of detailed information on what the increased government spending will be used for, it is difficult to comment on how meeting the 9 percent target will affect individual budget items at central and local levels. To help achieve the national policy of free MCH, extra financing could be allocated centrally or provincially to individual facilities on a utilization-linked basis. Provincial spending could also be increased through budget norms that are currently being developed. Figure 17: Planned health expenditure versus the 9% target, FY 2011/12 Planned health expenditure vs 9% target, FY 2011/12 Total Recurrent Central Sub-national Total Central Sub-national Total Source: State Budget Plan 33. Around the provinces there are large differences in the share of total spending allocated to health. The most populated provinces are those which have the largest gap to bridge to meet the 9 percent health spending target, both for total and recurrent expenditure. Poorer and less populated provinces are closer to the 9 percent target (Figure 18). As discussed in more detail below, this may reflect differences in the priority given to health across provinces, but may also be due to differences in the costs of providing services due to geography, population density and other factors. As in the case of central spending, the gap varies depending on whether one considers total (including foreign capital) or recurrent expenditure only. Figure 18: Planned health expenditure versus the 9% target by province, FY 2011/12 Planned health expenditure vs 9% target by province, FY 2011/12 Total Recurrent Vientiane capital Savannakhet Savannakhet Champasack Loangprabang Loangprabang Vientiane province Khammouan Champasack Vientiane province Khammouan Vientiane capital Xayabouly Xayabouly Oudomxay Xiengkhuang Xiengkhuang Planned Huaphan Huaphan Oudomxay Borikhamxay Saravanh Attapeu Borikhamxay Saravanh Loangnamtha Loangnamtha Gap to 9% Phongsaly Phongsaly Attapeu Sekong Bokeo Bokeo Sekong 0 20,000 40,000 0 10,000 20,000 LAK, millions LAK, millions Source: State Budget Plan Overcoming Provincial Disparities in Government Health Spending 34. The Ministry of Finance has a role in approving sub-national budget allocations and according to the Budget Law and regulations, budget norms should be developed for all types of expenditure. Both the Finance and the Health ministries are currently in the process of developing budget norms for non-wage expenditure33. (Additional information on the health sector budget process can be found in Appendix 5.) However, there are separate processes for allocating personnel resources (through the system of quota allocation applied by the Public Administration and Civil Service Authority [PACSA])34, and capital resources (through the investment budget). Details are provided in Table 3. Table 3: Existing allocation criteria for the government health budget Criteria for allocation Advantages Limitations Actions to be considered Capital Population criteria with Rational and The allocation of funds Update planning Investment allowance for equitable. for repairs, maintenance methods over time. Budget population and other non-salary Reflect changes in density/ distance from recurrent costs is too population distribution health facility. low relative to the level and transport links. of capital expenditure. Strengthen links between capital and recurrent costs budgeting. Personnel Quotas based on Distribution of Quota allocations to the To reduce vertical budget staffing standards and health care health sector have been inequity, consider forecasts of retirement workers by running at only 50-60% additional allowances for and staff turnover. headcount is of the annual bring public sector workers in reasonably equal level. The MOH needs a remote areas, in across provinces plan to cover attrition combination with other but less equal and fill all staffing norm human resource across districts, posts by 2020. strategies for recruiting local candidates at regional training centers.35 Non-salary The state budget Until 2005, the Because of the growth Introduce a mechanism recurr ent allocation system treats budget used to be of salaries and for equity in allocating budget non-wage recurrent sufficient to investment expenditure, non-personnel recurrent costs as a residual after finance utilities, the non-wage recurrent budget across provinces salaries, investment some supplies and budget has fallen and and districts.36 and other non- some drugs for the cannot now cover discretionary spending. poor. utilities or fuel costs. 35. Budget norms for non-salary recurrent costs could help alleviate inequalities related to the increasing reliance on user fees and RDF revenues for non-wage recurrent expenditures (Box 3). For health centers and district hospitals, the government could develop a budget norm that sets a target for the level of non-wage recurrent expenditure, or a target for the share of recurrent budget that should be allocated to non-wage recurrent costs. Targets could be based on essential running costs including utilities, maintenance and minor repairs, staff uniforms, basic medical supplies and subsidized essential drugs for the poorest patients. Box 3. Rationale for Budget Norms in the Health Sector The State Budget Law (No.02/NA 2006) calls for budget allocation norms to be developed for budget allocation targets for each local government area (province or district) and to each sector. According to the Implementing Decree (No.25/PM 2008) the Prime Minister shall issue decrees on budget norms, and the Minister of Finance shall issue detailed instructions on how budget norms are to be reflected in budget preparation. The main purposes for health sector budget norms are: (i) to improve equity within health budget allocation horizontally across districts and vertically between different levels - center, province, district; (ii) to align the allocation of provincial and district budgets across sectors, including health, with the goals and priorities of government strategy and policy. Norms are only one instrument for reducing inequality and increasing central leverage over sub-national administrations' efforts towards national health goals. In the Lao health sector the most important current levers for achieving equity in vertical and horizontal allocation are the planning processes for health facility investment, the MOH facilities standards and staffing norms, and the personnel quota allocation system. 36. The method for setting budget norms should not aim for equal per capita spending on non- wage recurrent expenditure across all districts or all provinces. As some district and provincial hospitals provide referral services, allocation of resources to districts and provinces has to take account of the type of hospital facilities located within their boundaries. Districts in particular show marked variation in health responsibilities, with type-A district hospitals providing surgery and type-B district hospitals not. There are also differences in health responsibilities across provinces. The most striking variations are seen in Vientiane Capital, and to some extent in Vientiane Province, where many residents use central hospitals for their primary and secondary health care. In addition, four provincial hospitalS37 will be upgraded and expanded to become regional hospitals, each providing a wider range of specialist services for neighboring provinces. Once these upgrades are fully implemented, these hospitals are likely to have more staff relative to population size, and a higher cost per hospitalization because of the greater intensity of treatment for more complex cases. However, potential for earning user fees is also likely to increase. Table 4 summarizes the options to consider in developing budget norms for non-wage recurrent costs and also predicts outcomes. Table 4: Summary of options to consider for the development of budget norms Options How? Expected outcome Consider the hospital's Develop norms to give higher levels or Reduce horizontal and capacity to earn revenues shares of budget allocations for facilities in vertical inequity, without from user fees, revolving poor and sparsely populated areas with reducing the incentive for drug funds, HEF, SASS, less capacity to generate user fees. facilities to collect revenues. SSO, free MCH payments and external grants. Clarify which types of Budget norms can clarify to provincial and Facilitate formulation and health facility expenditure district authorities what costs should be negotiation of PHO and DHO are under the responsibility financed from budget revenues, and what budgets. of the government (i.e. to be costs may be covered by user fees. funded from government budget), and which types of expenditure may be funded through user fees Provide additional In the short to medium term: develop Improve equity in budget resources for district and different budget norms for each type of allocation regional hospitals that facility. provide services to patients In the long term: social health insurance from outside their district will pay for patients who are referred to or province, hospitals in other districts or provinces. Set up mechanisms to Implement free MCH policy. Remove incentive that health prevent increases in user Set up lower user fees in sub-national facilities have to increase user fees and therefore, to facilities, fees and deny exemptions ensure affordable access for Monitor the percentage of patients who are from user fees. the poor provided with exemptions/free services. Increase public communication of user fees, free care and exemptions policies. Management of Revenues at Facility Level 37. The above sections have presented current weaknesses in the management of revenues at facility level. The Lao government is committed to improving the management of revenues and the financial sustainability of facilities. In this regard, a number of unresolved issues (as summarized in Table 5) need to be considered, along with options for reform. 38. In addition, there are important financial management capacity constraints that need to be addressed to improve revenue management at central and local levels. At present, most hospitals keep separate records for government budget, user fee revenues and RDF. However, due to capacity constraints, it remains difficult for MOH, MOF, social health insurance organizations and the national and sub-national administrations to obtain detailed facility-level information on the cost of providing health services. Ensuring comprehensive recording and reporting of these separate fund pools will be important for refining financing schemes and payment mechanisms. Table 5: Issues in the management of health sector revenues and options for reform Unresolved questions Issues Options for reform Should user fees be set on the Full cost recovery would mean very principle of "full cost steep increases in user fees, making recovery"? care unaffordable for many uninsured patients - not only the poor. How to counterbalance a If user fees increase, there is a risk Budget subsidies need to provide potential increase in user fees? that sub-national administrations will financial protection for the majority of the respond by reducing budget population who are uninsured, many of subsidies to their hospitals and health whom are poor. centers. What is the real cost of Hospital costs are not well-reported Improve management accounting and providing health care and audited. Management reporting systems at hospitals to obtain a services? How can the accounting and reporting of comprehensive report of hospital income financial situation of hospitals expenditures financed from and expenditure (see positive experience be assessed? government budget, user fees and of an FMIS pilot at the Mother & Child RDF are separated. Hospital). Simplify and unify the different regulations on how user fees and RDF revenues can be used. Can a single user-fee schedule Since 2006, there has been a single Differentiate the user-fee rates applied at be applied to all types and regulated user-fee schedule for all different levels of care. levels of hospitals? facilities although fees should vary Provide higher budget subsidies where according to the level of care, there are many uninsured poor people. remoteness, financial capacity of patients, etc. How to address hospitals' Hospitals have difficulty in Allow user-fee revenues to be passed concerns about the law's forecasting user-fee revenues for the through the single Treasury account requirements to include user- year. system into an account that is managed fee revenue in the general Hospitals are concerned about delays directly by the hospital director under revenues and budget? and line-item inflexibility if they can delegated authority. only use user-fee revenues based on Grant full flexibility for hospital directors annual budget procedures. to reallocate user-fee revenues between Returning unspent revenues to the non-wage budget line-items. MOF removes the incentive to Let hospitals keep unspent user-fee increase utilization of health care revenues for the following fiscal year. services. How to pay for bonuses and There is no policy guidance for a Set clear regulations on incentives and performance incentives at transparent bonus or incentive bonuses taking into account objectives of hospitals? payment system. quality, efficiency, rational use of drugs Unintended negative effects can and diagnostic tests, and equity. occur if hospitals pay staff a percentage of hospital profits, RDF profits or user fees (Appendix 4). Unresolved questions Issues Options for reform How to reform provider Fee-for-service payments and sale of As long as the majority of patients are payment methods? drugs for profit lead to inefficiency, uninsured, user fees based on capitation, unnecessary service provision, flat fees or case payments could be irrational prescribing and explored. uncontrolled growth of costs. How to improve drug The RDF decree and guidelines are i the short term: MOH should encourage procurement, distribution and not monitored or enforced. There is continuation of good practice in use? large variation in procurement and provinces with better management. pricing practices. In the medium to longer term: MOH should build on the reform experience in other regional countries such as Thailand or China (Appendix 4). 33 Recent updates from March 2012 based on publically disclosed documents from the World Bank Project on Public Financial Management Strengthening (P108787) indicate that (i) a decision is still pending on whether to formalize non-wage recurrent budget allocation norms for the health sector by PM decree, (ii) MoF used formulas for health budget allocation in FY 2011-2012, but these formulas were not formalized, and (iii) MPI continued to use the adopted system of capital budget norms for FY 2012-13 budget preparation. 34 The process for planning and allocating personnel, although equitable in principle, faces difficulties in implementation. PACSA allocates available quota for hiring across line ministries according to government plans. There are however vertical and horizontal inequities in allocation of the salary budget across provinces and districts, because most districts and some provinces face difficulty in recruiting high-level and mid-level health professionals. As a result, they fill vacancies with lower level staff. District health facilities in aggregate have excessive numbers of low level staff, and provincial facilities have excessive numbers of general practitioners and insufficient numbers of specialist doctors. 35 Recent analytic work by Capacity Plus has costed the incentives for attracting and retaining rural health workers based on the results of a Discrete Choice Experiment. Chee, G (2012). Costing of Incentives to Attract and Retain Rural Health Workers in Lao PDR. Abt Associates. 36 The debate has so far been concerned with the question of whether norms for provincial non-wage spending should be population- or facility-based. Given the focus on non-wage recurrent spending and the need to closely link spending with facility infrastructure and capacity to deliver services, these budget norms for non-wage spending should be facility based, perhaps involving fixed block grants or variable activity-based components. In the context of the government's national Free MCH policy, which pays facilities based on utilization of services, this would form a pillar for increasing non-wage spending at the local level. 37Champassak, Savannakhet, Luang Prabang, and Oudomxay. VI. Conclusion 39. This policy note on government spending for health in the Lao People's Democratic Republic (Lao PDR) has provided an overview of government budgetary health spending, primarily covering fiscal years (FY) 2005/06 to 2011/12. The note analyzes overall trends in government health financing and expenditure patterns and discusses some of the efficiency and equity issues pertaining to current government health spending patterns. The policy note is one of a series of health financing analyses, complementing earlier policy notes focusing on out-of-pocket spending as well as community-based and social health insurance schemes in the country. 40. Despite notable progress in health on some fronts, Lao PDR continues to have some of the worst maternal and child outcome indicators, both globally as well as in the East Asia and Pacific region. From a health financing perspective, out-of-pocket payments account for the largest share (46.4%) of total health expenditure in the country. This reliance on out-of- pocket payments represents a considerable financial barrier to utilization of health services. The prominence of out-of-pocket spending in the form of user fees and RDFs also raises concerns over management of funds at health facility level and regarding the potential for over prescription. In contrast, social health insurance expenditures are very low in Lao PDR: social insurance schemes cover about 11.4 percent of the population but account for only about 2.8 percent of total health spending. 41. By international standards, government spending on health - at 1.1 percent of GDP - is low and has been erratic over time, primarily due to external financing; domestically-financed government health spending has averaged about 0.5 percent of GDP in recent years. Health's share of the government budget was about 4.3 percent in Fiscal Year 2009/10, and has been increasing. The Lao government has committed to increasing government spending to 9 percent of the budget, implying roughly a three-fold rise compared to planned spending for Fiscal Year 2011/12. If the policy goal is to raise government health spending equitably across the provinces, achieving it will be challenging. Some provinces, notably the more heavily populated ones, will have large expenditure gaps to bridge. 42. There are large disparities in government health spending between central and sub-national levels. In Fiscal Year 2009/10, less than a third of government health spending occurred at the provincial level. Furthermore, provinces differ in the share of total government spending dedicated to health: provinces with higher levels of overall government spending per capital also tended to spend more on health. Per capita health spending was higher in more sparsely populated and poorer provinces such as Sekong and Attapeu, in contrast to Champassak, Savannakhet, and Vientiane Capital. Budgetary provincial spending rates, however, may not give an accurate picture of equity because of cross-provincial utilization, especially at higher levels of care. In addition, hospitals in more affluent provinces/cities are likely to raise a larger share of their review from user fees and drug sales. 43. Although the recently-announced planned increases in government health spending are welcome, challenges remain. These include ensuring that the additional resources are used to improve access to and utilization and quality of health services - especially in more remote areas - and progressively making additional domestically-financed resources available to reduce both dependence on external funding and out-of-pocket spending for health. To attain these objectives, the government should consider an appropriate mix of both demand-side and supply-side incentives. Instead of, or in addition to, setting a target for budgetary outlays for health, the government needs to improve the efficiency of existing outlays, the measurement of which requires the monitoring of key population health outputs. These should include focus on the level and equity of basic immunization rates, of skilled birth attendance, of institutional delivery rates, of need-based outpatient and inpatient utilization rates, and on adequate levels of financial protection from adverse health shocks. 44. The planned implementation of the free maternal and child health policy is a welcome step in the right direction. However, implementation will need to be complemented by improvements in the capacity of health facilities, not just in clinical and service availability terms, but also in terms of their ability to manage and allocate revenues appropriately. Current weaknesses include inconsistent implementation of user-fee regulations and revenue management, variation in management practices, weak procurement practices for drugs, and inadequate service provision levels. In addition, the planned removal of user fees, as envisioned under the free maternal and child health policy, may not be sufficient to improve utilization and inequalities across the country. To achieve this, the government should consider additional demand-side incentives, especially in rural areas. 45. Budget norms, if used prudently, can be a useful framework for determining allocations for all expenditure types. For capital investments, population distribution and needs should be considered. For recurrent costs, equity adjustments could help reduce vertical inequality in personnel budgets. The same principles could be applied to allocation of non-wage recurrent expenditure across provinces and districts, taking into consideration not just population size but capacity and health-sector needs. The government should also consider the capacity of health facilities to generate revenue through state health financing and insurance schemes, and provide clear guidance on which types of expenditure are to be funded by non-wage recurrent budgetary allocations and which are to be funded by user fees. Various unresolved questions remain, including on performance incentives at hospitals, on reforms to provider payment mechanisms, and on drug procurement practices. In conjunction with other reforms, use of informed budget norms will be necessary to achieve efficient and equitable use of resources as the government proceeds with plans to increase its spending on health to 9 percent. Appendix 1: Evolution of policy on user fees in Lao PDR since 199538 Table 6: User Fee Policy Evolution Decree 52/PM (6/1995) Revised Decree 52 (4/2005) Curative Law (11/2005) Instruction 2635/MOH (12/1995) Decree 381/PM (12/2005); Instruction 1646/MOF (7/2006); Decree 3 (3/2008) Relationship with First user-fee legislation Revised Decree 52 (4/2005) was submitted Supersedes Decree 52/PM previous to the government for endorsement but it legislation was recommended that instead the user-fee policy should be reflected in the Curative Law. Objective Not indicated [Revised Decree 52] Article 1: To promote [Curative Law] Article 1: To ensure equitable equal access to health care service for all; access to quality health services for all. To To implement fee exemption and non- protect the rights and interests of health care exemption policy; To collect revenues from professionals. To develop modern health care medical services services. Definition of user Not indicated [Revised Decree 52] Article 2: 2.1 Medical [Decree 381/PM] Article 2: Technical revenues fee services include examination, diagnosis, are a source of budget revenues collected by the and treatment of patients. 2.2 Free medical relevant government authorities from services service is covered by health insurance or ... to the public through the use of physical and social welfare. intellectual skills and materials, equipment, 2.4 Technical revenue is the revenue from machinery, properties and funds of the user fees for examination and treatment of government, including the use of the patients. 2.5 Non-technical revenue is from government's authorities in issuing certificates other sources e.g. cafeteria. and permits. Decree 52/PM (6/1995) Revised Decree 52 (4/2005) Curative Law (11/2005) Instruction 2635/MOH (12/1995) Decree 381/PM (12/2005); Instruction 1646/MOF (7/2006); Decree 3 (3/2008) Services coverage [Decree 52/PM] Article 6 listed seven items [Revised Decree 52] Article 4: There are 15 [Curative Law] Section V, Article 52 subject to user fees in central hospitals, items eligible for revenue collection 1. Fee for registration, health certificate and provincial hospitals and district hospitals [1. drugs, 2. laboratory 3. x-ray, 4. other medical certificate [1. medical records and other documents, 2. techniques, 5. equipment maintenance, 6. 2. Fee for room, food, ambulance supplies, 3. laboratory, radiology and other treatment (operations, obstetric procedures, 3. Fee for consultation, lab, X-ray, medicine, diagnostics, 4. medicine, 5. medical hemodialysis, rehabilitation, ...), 7. skilled medical supplies, surgery, drugs, medical appliances, 6. curative care costs (surgery, labor for difficult examinations and materials, and other technology rehabilitation, acupuncture), 7. room and treatment 8. artificial transplantation, 9. [Decree 381/pm] Article 7: Technical revenue board] room and board, 10. registration, 11. health can be charged for (1) training, analysis, survey Article 8: at health centers only a charge for certificate, 12. examination, 13. autopsy, 14. and information supply services; (2) fees for the medicine is permitted. ambulance, 15. funeral] signing or certifying documents, medical treatment fees, hospitalization fees ...; (3) fees collected on the use of professional equipment and use of premises; (4) sale of products using funds from the national budget; (5) supply of forms or documents... Fees and Exemptions Fee schedule Not available. Each hospital sets its own fee same as Original Decree 52 istruction 1646/MOF Article 8: provides a schedules. comprehensive list of service fees in health sector. MOH in coordination with MOF revised this fee schedule in 2008, reflected in Decree ______________ ______________________________3/2008. Process3for [Instruction 2635/MOH]: the process same as Original Decree 52 [Decree 381/PM] Article 11: The Ministry of setting fees should follow guidance from the Ministry Finance is entrusted to examine technical of Health. However, no guidance is revenue rates in coordination with the relevant available. Fee schedule proposed by authorities and to submit proposed rates to the hospital is subject to approval by Curative government for approval. Departmentt MOH. All hospitals may collect fees [central, provincial and districts]. In practice each hospital sets and adjusts its own fee schedule according to local economic situation and acceptability to patients. ____________________ ______________________ Decree 52/PM (6/1995) Revised Decree 52 (4/2005) Curative Law (11/2005) Instruction 2635/MOH (12/1995) Decree 381/PM (12/2005); Instruction 1646/MOF (7/2006); Decree 3 (3/2008) Fee payers [Decree 52/PM] Article 4: All citizens who [Revised Decree 52] Article 3 (3.6): Patients [Decree 381/PM] Article 3: Heads of business are not covered by exemptions under other than the targeted exempt population units, mass organizations, social organizations, Articles 1, 2 or 3 below. mentioned in 3.1 -3.5 (below) must pay user persons or juridical persons without fees, unless they are enrolled in a health differentiation by nationality carrying out insurance scheme; in this case the health activities in Lao PDR and using services insurance will pay the fees. provided by the government are liable to perform obligations to the national budget as stipulated in Article 7 of this Decree (Source of Technical Revenue). Exemption rules [Decree 52/PM] Article 1: Government [Revised Decree 52] Article 3 (3.1-3.5): [Curative Law ] Article 3(9) Free health care is employees and pensioners are exempted Government officials, retirees, handicapped provided for poor patients. Poor patients from user fee. persons, retired military officials, family of require a certificate issued by the relevant Article 2: Government employees' spouses military who died in line of duty, children authorities. and children are exempted from user fee. of retirees or civil servants under 18 years Article 3: Students, monks and the poor or old and spouses of retirees or civil servants the low income groups are exempt from who are not employed, students under 18, user fees. monks and novices are exempted from user Eligibility criteria or guidelines for fees. Social Insurance will cover their costs. identifying the poor are not available. Poor people with certification are exempt from fees. Implementation [Decree 52/PM] Article 10: For a low Not indicated [Curative Law] Poor/low income people need arrangement for income person, the hospital can accept to obtain a certificate from relevant authorities exemption affirmation by the Ministry of Labor and to obtain free health care service. Social Welfare (MOLSW), Social Welfare Provincial Office or the village head. [Instruction 2635/MOH] To obtain fee exemption, a low income person needs to provide a letter certified by village head; others such as civil servants, monks, students need to provide an ID from institutions they belong to. Decree 52/PM (6/1995) Revised Decree 52 (4/2005) Curative Law (11/2005) Instruction 2635/MOH (12/1995) Decree 381/PM (12/2005); Instruction 1646/MOF (7/2006); Decree 3 (3/2008) Subsidization for [Decree 52/PM] Article 1: For government [Revised Decree 52] Article 3 Social [Curative Law] Article 46-49: Different exemption employees, hospitals are reimbursed by the Insurance Fund for civil servants will insurance schemes will subsidize people who Social Security Department, MOLSW on a subsidize fees for civil servants, retired are members and pay contributions or fee for service basis. Article 13: The military officials and families (children premium. Government Welfare Funds will Ministry of Finance will allocate budget to under 18 years old), including families of subsidize fees for the poor who are unable to subsidize hospitals that provide services to military officials who died in duty, pay membership for any insurance schemes, government employees' spouses and students under 18 years old, monks and subject to regulations. children (Article 2) and students, monks novices. Government will reimburse and the poor (Article 3). The Ministry of hospitals for services to the poor. Health will propose the annual budget to the Ministry of Finance. [Instruction 2635/MOH Section V: Reimbursement for I exemption] I I Revenue Management Use of revenues [Decree 52/PM] Article 14: 20% of the [Revised Decree 52] Article 7: Revenue [Curative Law] Section V, Article 53 revenue from user fees shall be remitted to from user fee can be used for Using and monitoring of revenue from user fee general revenue of the government. The 1. Payment of equipment and drugs must be under the Budget Law. rest of the revenue can be used by hospitals necessary to maintain basic hospital [Decree 381/PM] Article 9: Ministries.., earning at their discretion, and if the revenue is operation 2. Minor renovation to improve technical revenues may apply such revenues to inadequate, the government will subsidize. quality of service. 3. Exempting the poor; annual budget expenditure lines: (i) salaries; (ii) There was confusion on whether to remit exempting pensioner and veterans during allowances to civil servants; (iii) recurrent 20% of revenue or 20% of profit. In some transitional period of Social Scheme expenditures; and (iv) development activities. cases the percentage was negotiable. development All revenue and expenditure must be recorded 4. Staff incentives and training as required by financial rules and revenues may not be used for other purposes inconsistently with the laws. [Decree 381/PM] Article 5: at the end of the fiscal year, any remaining technical revenues must be returned to the national budget, unless authorized to maintain as reserves. Decree 52/PM (6/1995) Revised Decree 52 (4/2005) Curative Law (11/2005) Instruction 2635/MOH (12/1995) Decree 381/PM (12/2005); Instruction 1646/MOF (7/2006); Decree 3 (3/2008) Procedures for [Instruction 2635/MOH] Section II [Decree 381/PM] Section II (Article 5-11) and managing indicates that revenues should be collected Chapter III (Article 12-14) Rules and revenues and kept in one place. The hospital should Management of Technical Revenues. Technical have an accounting system to manage the revenues can be used for ministries/agencies' revenue, and establish a treasury to recurrent expenditure items and need to be manage reserves and use of revenues recorded as required by financial rules. according to MOF procedures. Remaining revenues at the end of fiscal year must be returned to the national budget unless authorized. Finance authorities have the duty to manage technical revenues in coordination with line ministries, agencies to set the fee levels, monitor and control the implementation of revenues and expenditures, and approve reports. Ministries and agencies are responsible for submitting annual revenue collection and spending plan to MOF, submitting monthly quarterly and semi-annual and annual reports to MOF, collecting and applying technical revenues as authorized by MOF, proposing improvements to sources of revenues and the fee rates as appropriate to actual situation. Health Insurance Scheme Health insurance 1. Government Employee Scheme 1. Government Employee Scheme [Curative Law] Section V Financing (Article 43- scheme - Beneficiaries: civil servants, pensioners. 2. Veterans Scheme 53) - Fee For Services reimbursement model 3. Students Scheme 1. CBHI: voluntary contribution 2. Students, monks and the poor or low 4. Monks Scheme 2. SSAS: civil servants' contribution income exempt from user fees. MOF 5. Social Scheme for the poor 3. SHI: employer and employee contribution provides budget to hospitals for exemption 6. The rest must pay out of pocket 4. Private insurance: voluntary individual following the MOH plan. premium 5. Social Fund for the poor: contribution from government, individuals, community, domestic and international organizations. 38 Prior to 1995, public health care services were provided free of charge. Revolving Drug Funds were established in 1990 to provide simple health services in rural villages, ensure availability of drugs in public health facilities while waiving charges for the poorest. RDF policy was a predecessor of user fees and exemption policy. Appendix 2: Case study on drug prices and availability of drugs in the south (2011) In May 2011 a World Bank team conducted two field visits in the central and southern provinces. The first field visit was to health facilities in Khammouane and Savannakhet provinces while the second visit included facilities in Sekong and Champassak provinces. The primary objective of these field visits was to prepare the second World Bank Health Service Improvement Project. It therefore included provinces where the Health Service Improvement Project will be implemented (Sekong, Champassak and Savannakhet) and one province for comparison (Khammouane). During the field visits, the team asked for the list of drugs in all facilities visited and checked the availability of drugs. Data was collected from 14 facilities: 4 provincial hospitals, 5 district hospitals and 5 health centers.39 It is important to note that this assessment is only informative as facilities were not selected randomly and the facilities studied are those for which data could be accessed easily. The assessment reveals that in practice, the RDF decree and guidelines are not monitored or enforced. It shows wide variation in how RDFs are implemented across provincial health offices (PHOs), district health offices (DHOs), different types of hospital and health centers. Price disparities for the same drugs are higher in provinces with decentralized procurement, with the ratio of highest to lowest prices for the same product ranging from 2 to 1 to 13.8 to 1. Prices for the same drug are on average higher in provincial hospitals than in district hospitals and health centers (Table 7). Observations also revealed that only a minority of facilities post drug prices publicly. Table 7: Summary of drug availability and prices for essential drugs Drug name Drug Use Provincial District Health Centralized drug Decentralized drug hospital hospital center management management Price max/min Price max/min Price range (median) range price range price (median) ratio (median) ratio Amoxicillin Antibiotic 450-940 400-600 400-800 450-800 1.8:1 400-940 2.4:1 (550) (500) (500) (500) (600) Cotrimoxazole Antibiotic 115-204 300 250 204-250 1.2:1 115-300 2.6:1 (319) (227) (207) Paracetomol Pain killer 86-150 100-200 50-200 86-200 2.3:1 50-100 2:1 (97.5) (100) (100) (150) (100) Ferrous Anemia 22-100 0-100 (0) 0-50 (0) 0-100 (0) - 0-100 (35) - sulphate tablet (42.5) Albendezole/ Anthelmintic 850-3000 0-500 (0) 0 0 1:01 500-3000 6:1 mebendazole (1925) (850) Metronidazole Anti-fungal 85-500 200- 150-500 150-438 2.9:1 85-1000 11.8:1 (319) 1000 (200) (200) (500) (400) Atenelol/ High blood 340-1500 N/A N/A 1000 N/A 340-1500 4.4:1 Propanalol pressure (1000) (670) Hydro- High blood 500 500- N/A 500 N/A 500-1000 2:1 chlorthiazide pressure 1000 (750) (750) Furosemide heart failure/ 110-600 200 250- 145-1000 6.9:1 110-400 3.6:1 fluid (273) 1000 (600) (200) retention (625) Digoxin Heart failure 65-1000 500-900 N/A 1000 N/A 65-900 13.8:1 (80) (700) (290) Glibenclamide Diabetes 300 N/A N/A N/A N/A 300 N/A Type 2 Metformin Diabetes 250-1000 N/A N/A 1000 N/A 250-500 2:1 Type 2 (500) (375) Oxytocin Childbirth 4900-8000 3000- 2000- 5000- 0 2000-8000 4:1 (6453) 10000 10000 10000 (4450) (5000) (5000) (10000) Magnesium Pregnancy 500-5000 N/A N/A 500-4375 8.8:1 5000-6100 1.2:1 Sulfate (eclampsia) (4688) (2438) (5550) Oral Diarrhea 1300-2000 1300- 1000- 1375- 2.2:1 1000-2500 2.5:1 rehydration (1688) 2500 3000 3000 (1300) salts (2000) (1500) (2000) MEAN of 1296* 944* 908* 1520* 3:1 (1:1- 916* 4.1:1 (2:1- medians 8.8:1) 13.8:1) Source: Data collected in four southern provinces (May 2011) Note: Means calculated only for drugs available in provincial hospitals, district hospitals, health centers 40 The availability of drugs was also measured in Savannakhet and Khammouane provinces at provincial hospitals, district hospitals and health centers. The assessment revealed an overall low availability of medicines, especially for non-communicable diseases, in district hospitals and health centers. It also highlighted variation according to the drug considered and the level of care (Figure 19). Figure 19 : Availability of drugs in Savannakhet and Khammouane provinces (2011) 40 Availability of drugs in Savannakhet and Khammuane provinces, 2011 Albendezole/ mebendazole Amoxicillin Ateneloll Propanalol Average availability of drugs Cotnmoxazole Digoxin Ferrous sulphate fablet Furosemide Glibenclamide Hydro-chlorthiazide Magnesium Sulfate Metformin Metronidazole Oral rehydration salts Oxytocin Paracetomol 0 20 40 60 80 100 Percent (%) Health Center District Hospital Provincial Hospital Source: Data collected in the 4 Southern provinces, May 2011 The assessment also found a variety of practices in the use of RDF revenues, with some facilities using about 20 percent of funds as incentives to staff or others using RDF revenues for hiring contractors and paying honoraria to volunteers; in most cases however, RDF revenues are used as a flexible source for meeting non-wage recurrent costs. On a side note, the assessment found examples of private interests, such as public hospital staff selling drugs at home or at a private business, thus revealing weaknesses in management practices. 41 Appendix 3: Case study on local level revenues and expenditure in the health sector Overview of the case study As with Appendix 2, this case study builds on two field visits conducted in May 2011 in central and southern provinces. The first visit was to health facilities, PHOs and DHOs in Khammouane and Savannakhet while the second field visit included was to facilities in Sekong and Champassak provinces. The primary objective of these field visits was to prepare the second World Bank Health Service Improvement Project. It therefore included provinces where the project will be implemented (Sekong, Champassak and Savannakhet) and one province for comparison (Khammouane). During the field visits, the team systematically asked for the financial management statements of the institutions visited, for budgets received from different sources and for expenditure data. Financial data was collected for 25 institutions: three provincial hospitals, three provincial health offices and 19 districts.41 It is important to note that at the district level, budgets for district health offices and district hospitals are pooled. The same limitations as for Appendix 2 apply: this assessment can only be informative as facilities were not selected randomly and the availability of data was the main criteria for inclusion of the facility in the study. The sample is therefore not representative of the provinces visited. Objective of the case study The case study aims to present trends in the volume and composition of revenue and expenditure at the local level. It also aims to assess financial management capacities at provincial, district and health center levels. Sources of revenue at the local level At the local level health facilities rely on four major sources of revenue: the government budget for recurrent costs - this is part of the decentralized budget allocated to provinces by the Ministry of Finance; external assistance in the form of donor support or projects; revolving drug funds collected through the sale of drugs to patients at facilities; and technical revenues, also known as user fees, which are collected by facilities when providing services to patients. 42 Finding 1: The financing mix at local level is uneven Sources of revenue vary significantly from one facility to another. Khammouane and Champassak provincial hospitals present similar patterns: about a third of total revenues come from the government's budget, a third from technical revenues and the rest from revolving drug funds and projects. In Sekong facilities rely more heavily on the government budget and, in the case of districts, on external financing. Finally, provincial health offices do not collect revenues from user fees or RDF and external resources represent their main financing source, accounting for over 75 percent of funds in Sekong and Champassak, and 50 percent in Savannakhet. External assistance is therefore recorded at the provincial health office level rather than at the provincial hospital level (Figure 20). Figure 20: Health facility sources of revenue, FY200911 Health facility sources of revenue, FY 2009/11 Lamam district (Sekong) PH Champasak PH Khammuane PH Sekong PHO Champasak PHO Savannakhet PHO Sekong Thateng district (Sekong) 0 20 40 60 80 100 Percent (%) Govt.budget Technical revenues DRF Projet/NGO Source: Data collected in health facilities, DHO and PHO, May 2011 Besides variations in the financing mix, there are also variations in the overall budget received by each institution. In our sample total budget varied from KN 614 million in Lamam district to KN 29,709 million at Savannakhet PHO. The latter receives about KN 15,000 million of external assistance, from 41 different projects (Table 8). Table 8: Health facility revenues by source (2009/10) (million kip) Govt. Technical RDF Project/ Total budget revenues NGO PH Khammouane 3,100 2,178 2,741 135 8,155 PH Champassak 4,857 4,484 6,667 - 16,008 PH Sekong 2,155 432 439 - 3,026 PHO Savannakhet 14,778 - - 14,931 29,709 PHO Champassak 2,162 - - 7,347 9,510 PHO Sekong 1,886 - - 6,522 8,408 Lamam district 349 - 6 259 614 Thateng district 723 24 56 450 1,253 Source: Data collected in health facilities, DHO and PHO (May 2011) 43 Finding 2: The government budget almost exclusively finances wages Most government budget to the provinces is allocated to two human resource categories: salaries and compensation/allowances. The remainder is allocated to operating costs, which amount to about 10-15 percent of government budget at provincial hospitals and about 5 percent at district level (Table 9).42 The government budget flows from the Ministry of Finance to the provincial level. Provinces then allocate budget on the basis of need. It is unclear how needs are assessed and analysis of per capita spending from the government shows a wide variation: from KN 3,500 per capita in Kaisone district to about KN 20,000 in Thateng district; from KN 7,000 per capita at Champassak Provincial Hospital to 21,500 at Sekong Provincial Hospital (Table 9). Table 9: Health facility expenditure from government budget, FY 2009/10 Institution or name of district Government budget Government budget Government budget on (kip per capita) on salaries (%) salaries and compensation Champassak Province Provincial Hospital 7,232 51% 88% Provincial Health Office 3,220 59% 61% Bachieng 9,962 89% 99% Khammouane Province Provincial Hospital 8,227 49% 81% Xebangfai 25,016 56% 87% Savannakhet Province Provincial Health Office 16,050 69% 97% Nong 14,064 60% 87% Atsaphanthong 17,316 69% 95% Kaisone 3,488 79% 93% Outhoumphone 7,691 72% 95% Phin 9,853 70% 94% Sepon 14,197 71% 95% Thapangthong 11,587 61% 92% Songkhone 10,438 72% 96% Champhone 10,575 70% 97% Atsaphone 8,988 63% 93% Vilabouly 12,487 65% 93% Xaibouly 9,792 72% 95% Xaiphouthong 10,409 65% 94% Phalanxay 9,739 57% 92% Sonbuly 8,539 67% 94% Sekong Province Provincial Hospital 21,507 39% 92% Provincial Health Office 18,821 58% 64% Lamam 10,826 65% 82% Thateng 19,931 52% 89% Source: Data collected in health facilities, DHO and PHO (May 2011) 44 Note: Per capita government budget was estimated using census projections for 2010 (provincial population for provincial hospitals and PHOs; district population for district hospitals) Interviews with key informants revealed that DHOs and health facilities endure long delays before receiving the budget from the province. As a result, salaries are paid every two or three months and the implementation of some activities is postponed (e.g. in Lamam and Kaleum). When facilities do receive the government budget, priority is given to the payment of salaries. As there is often a gap between planned budget and outturn, facilities fall short in the payment of some recurrent costs. Finding 3: Facilities have to rely on technical revenues for non-wage recurrent expenditure At the local level, facilities collect technical revenues from several sources (Figure 21) and the respective share of each source varies from one facility to another. In facilities where Health Equity Funds (HEFs) are implemented, HEF revenues account for more than half of revenues collected at the facility (62 percent at Sekong Provincial Hospital) and the rest comes from the delivery of services. At other facilities (Champassak and Sekong Provincial Hospitals), the largest sources of revenues are laboratory services and health care services. In the three provincial hospitals in the study, SSO and SASS revenues account for less than 1 percent of the total technical revenues collected. CBHI revenues are also very limited (4 percent at Sekong Provincial Hospital43). Figure 21: Sources of technical revenue at provincial hospitals, FY 2009/10 Sources of technical revenues at provincial hospitals, FY 2009/10 0 0 0 a- Champasak PH Khammuane PH Sekong PH IPD OPD Laboratory Emergency+ICU Specialist services HEF Others Source: Data collected in provincial hospitals, May 2011 Given the low level of government spending on operating costs, health facilities must rely on technical revenues for some major recurrent costs such as electricity, water or petrol, and for maintenance, equipment and capital expenditure (Figure 22). The assessment revealed that technical revenues collected by health facilities are almost all allocated to operating costs: they represent more than 90 percent of expenditure out of technical revenues in Champassak and Khammouane provincial hospitals. 45 Figure 22: Spending on operational costs from government budget and technical revenue Operating costs out of technical revenues, FY 2009/10 Operating costs out of governments budget, FY 2009/10 Atsaphanthong district Bachieng district Champasak PH Champasak PH Champasak PHO Khammuane PH Lamam district Nong district Savannakhet PHO Sebangfai district Khammuane PH Sek.ng PH Sekong PH Sekong PHO Thateng district 5,000 10,000 400 LAK, millions LAK, millions S purchs,norequIpme tniunnt unirm g g~ Punnoeupetnluat - unfo - oth- Source: Data collected in health facilities, DHO and PHO, May 2011 Finding 4: Localfacilities do not implement user fee regulations Provinces and districts are implementing a mixture of new user fees set out in Decree 3/2008 and the old Decree 52. Hospitals use the new fee schedule for items if they think the new rate high enough to cover the relevant costs. For some procedures, however, such as normal child- delivery and caesarean section, the fees set in Decree 3 are felt to be too low (lower in fact than the MOH approved payment rates for free MCH pilots), and hospitals therefore continue to set their own rates. The new user fee decree has not yet been implemented in Health Centers. Provinces are still implementing Decree 52 provisions requiring 20% (sometimes more) of user fees to be paid to the Treasury, although this is no longer required. They are not yet implementing the new State Budget Law provisions requiring that expenditure financed by technical revenues be included in the budget, although two provinces prepared budget plans for technical revenues to be submitted to the Treasury for 2011-12. User fee revenue is currently retained and managed directly by the hospital director together with RDF revenues, usually in the hospital's own bank account. Technical revenues are generally used for recurrent costs: contractual staff wages; allowances; cleaning; electricity; water; telecommunications; maintenance of medical equipment; vehicles; buildings; office supplies; receptions; and transport. 46 Finding 5: There is poor management of the different health financing schemes at local level The HEF, SASS, SSO and CBHI contribute additional sources of revenue at facilities. However, high transaction costs and complications with cost-recovery are involved. Health equity fund revenues are not pooled with other technical revenues as facilities have to adhere to HEF guidelines for the use of HEF surplus: 50 percent for staff incentives; 50 percent for administration (Kaleum and Soukhouma). The SASS is a relatively new scheme in the southern provinces and all facilities report a deficit in using it. The share of contributions to be divided between provincial and district hospitals vary by the type of district hospital: 80 percent would go to the district hospital if it is a type-A district hospital, and 40 percent in the case of type-B district hospitals (which do not offer surgery but refer cases to provincial hospitals). Similarly, many facilities in Champassak and Savannakhet experience a deficit with the CBHI as expenditures exceed members' contributions. However, in districts of Sekong where HEFs are running, there is no deficit for CBHI as the HEF pays the contribution of CBHI members. The financial situation is generally better for the SSO, with facilities generating a surplus that is used to bridge the deficit from SASS and CBHI. Finding 6: Inefficiencies in the procurement of drugs at the local level prevent effective management of revolving drugfunds The assessment revealed a wide range of RDF management practices, both good and poor. In two of the four provinces, drug procurement for district hospitals and health centers is centralized and mainly follows the Ministry of Health's RDF guidelines (Table 10). In contrast, in all four provinces the provincial hospitals procure drugs by themselves, either quarterly or monthly. Purchase is made directly from suppliers after some price comparison, but without PHO control. Suppliers typically provide one-to-three months of credit and on request will replace drugs with new stocks when they are six months from their expiry date. Supplier prices include transport costs to the province or district. One province and most districts appear to buy from multiple suppliers based on price. Other provinces buy from a single supplier. In one case, the province's supply has been monopolized by a single supplier, to which the province owes high debts. 47 Table 10: Observations on drug procurement practices Use of budget funds Use of RDF Centralized procurement Provincial Food and Drug DHs carry out procurement for Department prepares the the RDFs of the district hospitals procurement plan and conducts and HCs, overseen by the district bidding, overseen by the RDF committee. provincial RDF committee. PHO monitors district RDFs and The provincial RDF committee DHOs supply and monitor HC certifies receipt of drugs. drug management. Decentralized procurement HCs order their own drugs from pharmacies in the area, generally do not compare prices, and pay by sending cash via bus driver. HCs have to send RDF reports monthly to DHOs, but in one district reporting and procurement had not occurred for 5-6 months. HCs have to meet their own transport costs to collect supplies from DHO or supplier. For remote HCs, these costs are high and selling price of drugs varies across HCs and can be above market price. Prescribing practices are not optimal: many doctors prescribe by brand name rather than generic name, arguing that staff in hospital pharmacies do not know the generic names. This can reflect a conflict of interests: in one hospital, doctors who also work for a direct sale company prescribe drugs not available in the hospital, but instead sell the medicines from their homes. In another hospital, drugs are prescribed and then sold at the doctors' private clinics. Finding 7: Irrational mark-up practices result in high drug prices Mark-ups were observed to range from 10-50 percent. Variation mainly arises from rounding up and from comparison with market prices. One province has even higher mark-ups and above-market prices because of its high debts to suppliers. Provinces differ in how they divide the permitted mark-up of 25 percent between the entity that procures the drugs at provincial or district level, the facility (district hospital, health center) that sell the drugs, and the fund for procurement of additional drugs for the poor. The two provinces with centralized procurement have a clear division of the mark-up. Others allow hospitals and health centers to make their own decisions on adding mark-ups, and have no clear allocation of funds for procurement of drugs for the poor. Finding 8: Due to poor management, health facilities do not benefit from the RDF surplus In all facilities, the surplus from RDFs is limited as the largest share of RDF revenues is used to replenish drug stocks. As RDF revenues and technical revenues are not always kept separately, facilities cannot always identify their RDF surplus. In Savannakhet and Champassak, when RDF and technical revenues are pooled, the RDF surplus is used for contractual staff wages, health volunteer allowances and operating costs. In Khammouane and Sekong, where RDF revenues are kept in a special account, the surplus is used for maintenance of the pharmacy and administrative costs related to RDF. 48 The assessment also revealed that some health centers are unable to retain the surplus generated from the sale of drugs and use all revenues collected for drug replenishment. Finding 9: Donor/project funds vary significantly by province and are difficult to manage at local level Health facilities generally keep track of external support received but they are not able to provide detailed information on the use of such resources. Projects vary in their scope and scale, from small projects that provide equipment, to large projects such as the Health Service Improvement Project, providing financial support for recurrent costs and activities. Health facilities also receive support from vertical programs such as support for malaria from the Global Fund. The number of projects implemented also varies by province: the provincial health office of Sekong reported 15 projects in 2009/10 against 41 reported by the Savannakhet provincial health office. Although it is difficult to estimate the exact share of facilities' resources that comes from donor/project funds, there is evidence that this share is important. Management of large resources can be difficult at facility level, providing an additional workload for staff. As a result, the execution rate for external financing is lower than it is for government budget funds. In addition, as the amount of external financing arriving each year is unpredictable, dependence on such funds raises challenges in terms of financial sustainability. 39 In Khammouane province: provincial hospital, 2 district hospitals, 2 health centers. In Savannakhet province: provincial hospital 2 district hospitals, 2 health centers. In Champassak province: provincial hospital, 1 district hospital. In Sekong province: provincial hospital, 1 health center. 40 The sample includes the provincial hospital, 2 district hospitals and 2 health centers in Khammouane province as well as the provincial hospital, 2 district hospitals and 2 health centers in Savannakhet province. 41 In Champassak Province: Provincial Hospital, Provincial Health Office, 1 district (Bachieng). In Khammouane province: Provincial Hospital, 1 district (Xebangfai). In Savannakhet province: Provincial Health Office, 15 districts (Nong, Atsaphanthong, Kaisone, Outhoumphone, Phin, Sepon, Thapangthong, Songkhone, Champhone, Atsaphone, Vilabouly, Xaibouly, Xaiphouthong, Phalanxay, Sonbuly). In Sekong province: Provincial Hospital, Provincial Health Office, 2 districts (Lamam, Thateng). 42 A small amounts also goes to compensations and allowances; new purchasing for operations and capital investment (only at Sekong PHO). 43 At Sekong Provincial Hospital, the CBHI scheme is subsidized by HEFs. 49 Appendix 4: China's Reforms of Drug Sales and Private Revenues in Public Hospitals China introduced partial autonomy in its public hospitals after 1980, with the goals of reducing the government's financial burden while expanding and upgrading hospital infrastructure. Typically most public hospitals use a hybrid of public hospital rules (for government funds) and market-oriented hospital rules for user fee income and drug sales. Most public hospitals charge social insurance funds and patients on a fee-for-service basis. Alongside hospital reform, there has been scaling-up of social health insurance coverage. Based on studies and lessons from these earlier reforms, China has recently been seeking to identify what works best and is introducing a new phase of reform to strengthen the public welfare functions of hospitals, and to control the tendency for hospitals and doctors to pursue profit. Hospital Reform: 1980-2005 Hospital budget subsidies were in decline and covered only basic salaries. In response, hospitals were then allowed to: * Charge higher prices for higher quality services (while a regulated fee schedule sets prices below cost for basic services); * Pay staff incentives from profits; and * Enter into profit-sharing private partnerships for new equipment and facilities. Positive Achievements of These Reforms This reform led to: * Increased number of hospitals; * Increased volume of hospital care; and * Experiments with alternative models of governance, management, and provider payment. Unintended Adverse Effects Unintended adverse effects were also observed: * The growth of hospitals was unequal: higher level urban hospitals grew at the expense of more cost-effective primary care and outpatient care; * Rapid growth of tertiary hospital expenditure with deficit of social insurance funds in some places; * Decline in hospital efficiency; * Private partnerships to supply equipment under profit-share agreements led to non- standard procedures forbidden by the government; * Poor and uninsured patients had difficulty accessing higher-level hospitals for complex treatments because the hospitals were crowded with patients suffering minor conditions; * Irrational over-provision of drugs and high-tech services for more affluent patients; * Conflict and mistrust between hospitals and patients, as costs continuously increased. Adjustment to hospital reform: 2006-2010: "Constructing a Socialistic Harmonious Society" China's new direction is based on strengthening the public welfare function of public hospitals and health services. The main elements of the new direction are: * Strengthening the role of public revenues (budget and social health insurance); * Strengthening government input in planning and supervision; * Improving management and quality of services; * Promoting efficiency; * Reducing patient expenses; * Separating governance and management; strengthening governance bodies; * Separating the non-profit and for-profit activities of hospitals and other health service providers, including moving the for-profit sale of drugs out of hospitals; * Reforming payment methods to eliminate incentives for irrational over-provision of services; and * Improving compensation and incentive policies to motivate staff in line with public welfare objectives rather than profit-making. 51 Appendix 5: Budget Process in the Health Sector as determined during the 2008 Public Expenditure Tracking Survey4 Budget preparation starts at the district level, as health centers (HCs) only make informal requests for resources but are not actively involved in formal budget proposal procedures. Most HCs do not prepare or submit budget proposals to the District Health Office (DHO) or discuss the annual budget with DHOs. Links between HCs and DHOs in the budget process are weak. DHOs receive guidance and instructions from Provincial Health Offices (PHO) on budget planning and use estimates of current year funding needs, expenditure from the previous fiscal year, and the proposed budget of the previous fiscal year as criteria to decide the total amount of annual budget proposal. However, DHOs understand that the approved budget is likely to be different from what is proposed. DHOs prepare annual budget plans and submit proposals to the District Finance Office (DFO) for approval and to PHOs for official acknowledgement. PHOs consolidate budgets received from the DHOs without adjusting the proposed budgets and then submit budget proposals to the Provincial Finance Office (PFO). PHOs also submit budget proposals to the MOH, although not for approval. PFOs receive health budget proposals from PHOs and DFOs, and submit provincial budget proposals, which include health budget proposals, to the MOF. When PFOs get budget approvals from the MOF, they approve budgets to PHO according to what the MOF accepts. The budgets are then approved to the DFOs, which in turn approve budgets to the DHOs. However, due to the often-large difference in approved and proposed budgets, after receiving approval of annual budget by PFO, PHO officers submit quarterly budget proposals to request non-salary funds. National Assembly Appra1 -- - Sub a PF PHO 4 World Bank (2008). Lao PDR Public Expenditure Tracking Survey in Primary Education and Primary Health. Vientiane. March 2008; Diagram taken from Asian Development Bank (2009). Technical Assistance Consultant's Report PPTA 7167-LAO: Health Sector Development Program Project. June 2009. 52